Histogram equalization using java

Question

I want to perform histogram equalization for an RGB image. For this , at first I generate the histogram using the following steps :

1)Map the intensity in the range [0,255] , by taking the gray value as val=img.getRGB(j, i) & 0xFF;

2)count the number of pixels corresponding to each intensity value(0-255)

3)Plot the histogram.

4)perform equalization

5)Now the problem I am concerned with is , to map to the RGB image corresponding to the equalized histogram. How do I do that? All are grayvalues .Any solution?

Can you show us any code you've written so far that talks about the steps you completed? Also, if you use each combination of RGB values as a single bin in your histogram, you will not be able to store this histogram in memory because you will require an array of 256^3 x 64 bytes (assuming integer), and unfortunately this is too fine to have any discriminative power. It's best if you quantize the bins before computing the histogram. — rayryeng

Settembrini Settembrini · Accepted Answer · 2015-08-14T22:40:48

I did this once in Java. Input is a grayvalue buffered Image bi. Histogram of bi will be equalized. Here is the code.

            int width =bi.getWidth();
            int height =bi.getHeight();
            int anzpixel= width*height;
            int[] histogram = new int[255];
            int[] iarray = new int[1];
            int i =0;

            //read pixel intensities into histogram
            for (int x = 1; x < width; x++) {
                for (int y = 1; y < height; y++) {
                    int valueBefore=bi.getRaster().getPixel(x, y,iarray)[0];
                    histogram[valueBefore]++;
                }
            }

            int sum =0;
         // build a Lookup table LUT containing scale factor
            float[] lut = new float[anzpixel];
            for ( i=0; i < 255; ++i )
            {
                sum += histogram[i];
                lut[i] = sum * 255 / anzpixel;
            }

            // transform image using sum histogram as a Lookup table
            for (int x = 1; x < width; x++) {
                for (int y = 1; y < height; y++) {
                    int valueBefore=bi.getRaster().getPixel(x, y,iarray)[0];
                    int valueAfter= (int) lut[valueBefore];
                    iarray[0]=valueAfter;
                     bi.getRaster().setPixel(x, y, iarray); 
                }
            }

Histogram equalization using java

1 Answers